dld-kit/dld
Decision-Linked Development (DLD) — a workflow for recording, linking, and maintaining development decisions alongside code. Skills for planning, recording, implementing, auditing, and documenting decisions via @decision annotations.
68
Quality
68%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
dld-retrofit
skills/dld-retrofit/SKILL.md
Discovery
67%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
The description effectively communicates specific capabilities with concrete actions and has a clear niche that distinguishes it from other skills. However, it relies on domain-specific jargon ('DLD') without explanation and lacks an explicit 'Use when...' clause that would help Claude know when to select this skill.
Suggestions
Add a 'Use when...' clause with trigger terms like 'bootstrap decisions', 'infer architecture decisions', 'document existing code decisions', or 'create decision records from code'
Expand or clarify 'DLD' acronym and include natural variations users might say such as 'architecture decisions', 'design decisions', or 'ADR'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'Analyzes code to infer rationale', 'generates decision records', and 'adds @decision annotations'. These are clear, actionable capabilities. | 3 / 3 |
Completeness | Clearly answers 'what' (analyzes code, generates decision records, adds annotations) but lacks an explicit 'Use when...' clause. The 'when' is only implied through the 'Bootstrap' context. | 2 / 3 |
Trigger Term Quality | Contains some relevant terms like 'DLD decisions', 'codebase', 'decision records', '@decision annotations', but 'DLD' is jargon that users may not naturally say. Missing common variations like 'architecture decisions', 'ADR', or 'design decisions'. | 2 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with specific domain focus on DLD decisions, decision records, and @decision annotations. Unlikely to conflict with general code analysis or documentation skills due to the specialized terminology. | 3 / 3 |
Total | 10 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured skill with excellent actionability and workflow clarity. The step-by-step process is clear with proper user checkpoints and executable commands. The main weakness is moderate verbosity in explanatory sections that could be tightened, and the content could benefit from splitting some detailed guidance into reference files.
Suggestions
Tighten the mode descriptions in Step 3 - Claude can infer the implications of 'file-level' vs 'method-level' references without the extended explanations
Consider moving the detailed 'System-framing decisions' and 'Implementation-level decisions' guidance to a separate RETROFIT-PATTERNS.md reference file to reduce the main skill's length
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is reasonably efficient but includes some verbose explanations that could be tightened, such as the detailed mode descriptions and repeated explanations of what decisions should capture. The interaction prompts are well-structured but lengthy. | 2 / 3 |
Actionability | Provides concrete, executable bash commands with proper variable usage, clear script paths, and specific examples of decision record format. The workflow steps are copy-paste ready with actual command syntax. | 3 / 3 |
Workflow Clarity | Excellent multi-step workflow with clear sequencing (Steps 1-9), explicit checkpoints (user approval before proceeding), and logical progression from analysis through generation to summary. Prerequisites are clearly stated upfront. | 3 / 3 |
Progressive Disclosure | Content is well-organized with clear sections, but the skill is somewhat monolithic. References to related skills (/dld-init, /dld-decide, /dld-snapshot) are mentioned but the detailed mode/broad mode explanations and decision identification guidance could potentially be split into separate reference files. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.